{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:E3RI5LYPLORKORFO33HRPSIJJN","short_pith_number":"pith:E3RI5LYP","schema_version":"1.0","canonical_sha256":"26e28eaf0f5ba2a744aedecf17c9094b4c7798b314c1b642b1d0a3dbfd1d5be1","source":{"kind":"arxiv","id":"2306.08189","version":1},"attestation_state":"computed","paper":{"title":"Language models are not naysayers: An analysis of language models on negation benchmarks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Karin Verspoor, Thinh Hung Truong, Timothy Baldwin, Trevor Cohn","submitted_at":"2023-06-14T01:16:37Z","abstract_excerpt":"Negation has been shown to be a major bottleneck for masked language models, such as BERT. However, whether this finding still holds for larger-sized auto-regressive language models (``LLMs'') has not been studied comprehensively. With the ever-increasing volume of research and applications of LLMs, we take a step back to evaluate the ability of current-generation LLMs to handle negation, a fundamental linguistic phenomenon that is central to language understanding. We evaluate different LLMs -- including the open-source GPT-neo, GPT-3, and InstructGPT -- against a wide range of negation bench"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2306.08189","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-14T01:16:37Z","cross_cats_sorted":[],"title_canon_sha256":"bcc1a684b005bf632458308b4e927c6010eddd341d3a34c0f82747f1cb6b1212","abstract_canon_sha256":"e40ba8fb6b0fac493381c9bead8758b498810ddfbc08b9e04382c6c9560cd2a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:20:49.125248Z","signature_b64":"wp1+N+NJNGNIvB7F7FXd8qiD4anw58mzRP2A/NJWQSMqo5yfIgf3D2V2DXm9QckOPf+G2KHgFUvP0zD/TVuHAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"26e28eaf0f5ba2a744aedecf17c9094b4c7798b314c1b642b1d0a3dbfd1d5be1","last_reissued_at":"2026-07-05T06:20:49.124868Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:20:49.124868Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Language models are not naysayers: An analysis of language models on negation benchmarks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Karin Verspoor, Thinh Hung Truong, Timothy Baldwin, Trevor Cohn","submitted_at":"2023-06-14T01:16:37Z","abstract_excerpt":"Negation has been shown to be a major bottleneck for masked language models, such as BERT. However, whether this finding still holds for larger-sized auto-regressive language models (``LLMs'') has not been studied comprehensively. With the ever-increasing volume of research and applications of LLMs, we take a step back to evaluate the ability of current-generation LLMs to handle negation, a fundamental linguistic phenomenon that is central to language understanding. We evaluate different LLMs -- including the open-source GPT-neo, GPT-3, and InstructGPT -- against a wide range of negation bench"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.08189","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2306.08189/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2306.08189","created_at":"2026-07-05T06:20:49.124930+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.08189v1","created_at":"2026-07-05T06:20:49.124930+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.08189","created_at":"2026-07-05T06:20:49.124930+00:00"},{"alias_kind":"pith_short_12","alias_value":"E3RI5LYPLORK","created_at":"2026-07-05T06:20:49.124930+00:00"},{"alias_kind":"pith_short_16","alias_value":"E3RI5LYPLORKORFO","created_at":"2026-07-05T06:20:49.124930+00:00"},{"alias_kind":"pith_short_8","alias_value":"E3RI5LYP","created_at":"2026-07-05T06:20:49.124930+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.16697","citing_title":"QueryBandits for Hallucination Mitigation: Exploiting Semantic Features for No-Regret Rewriting","ref_index":105,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E3RI5LYPLORKORFO33HRPSIJJN","json":"https://pith.science/pith/E3RI5LYPLORKORFO33HRPSIJJN.json","graph_json":"https://pith.science/api/pith-number/E3RI5LYPLORKORFO33HRPSIJJN/graph.json","events_json":"https://pith.science/api/pith-number/E3RI5LYPLORKORFO33HRPSIJJN/events.json","paper":"https://pith.science/paper/E3RI5LYP"},"agent_actions":{"view_html":"https://pith.science/pith/E3RI5LYPLORKORFO33HRPSIJJN","download_json":"https://pith.science/pith/E3RI5LYPLORKORFO33HRPSIJJN.json","view_paper":"https://pith.science/paper/E3RI5LYP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.08189&json=true","fetch_graph":"https://pith.science/api/pith-number/E3RI5LYPLORKORFO33HRPSIJJN/graph.json","fetch_events":"https://pith.science/api/pith-number/E3RI5LYPLORKORFO33HRPSIJJN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E3RI5LYPLORKORFO33HRPSIJJN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E3RI5LYPLORKORFO33HRPSIJJN/action/storage_attestation","attest_author":"https://pith.science/pith/E3RI5LYPLORKORFO33HRPSIJJN/action/author_attestation","sign_citation":"https://pith.science/pith/E3RI5LYPLORKORFO33HRPSIJJN/action/citation_signature","submit_replication":"https://pith.science/pith/E3RI5LYPLORKORFO33HRPSIJJN/action/replication_record"}},"created_at":"2026-07-05T06:20:49.124930+00:00","updated_at":"2026-07-05T06:20:49.124930+00:00"}